dripelectronics
Well-known member
ah thats really cool , :thumb: nice job.
g.
g.
T4b ownage
- Thread starter dripelectronics
- Start date
Help Support GroupDIY Audio Forum:
[quote author="dripelectronics"]ah thats really cool , :thumb: nice job.
g.[/quote]
Thanks. I noticed from one of your posts you're planning on using a National Instruments Data Acquisition board for your tester. Mine is also based around one of their boards, although a much older version, obviously. I think you'll find them very easy to work with.
g.[/quote]
Thanks. I noticed from one of your posts you're planning on using a National Instruments Data Acquisition board for your tester. Mine is also based around one of their boards, although a much older version, obviously. I think you'll find them very easy to work with.
CroatianSensation
Well-known member
This is awesome!!!
I'm in for at least three!
My Drip2a's are getting close to being done.
All I need now are rack cases and some T4's.
Excellent work!
:guinness: + :guinness: = :grin:
I'm in for at least three!
My Drip2a's are getting close to being done.
All I need now are rack cases and some T4's.
Excellent work!
:guinness: + :guinness: = :grin:
Infernal_Death
Well-known member
Very cool. Depending on the final price i might be in for 1 piece.
Flo
Flo
dripelectronics
Well-known member
made some progress using the nsl 5910's
i think i should be able to work this out faily quickly.
these are the light to dark resistance curves of a batch of 30 or
so LDR's. (light dependant resistors)
interesting enough the dark to light curves seem inconsequential
for they are in the millisecond range ,
super fast ,
makes me believe that the el panel has more to do with that
then one thinks.
infact i would say the panel is a major player in this equation ,
to the point of being able to be a little loose on cell matching.\
the Brand X ldrs both land up on the red line
the UREI ldrs are more quick
alot of the LDRs are ending up more like the urei then i expected
almost dead on .
the brand x are harder to come by.
as with the 'smooth' cells
the intresing thing is now im learning attack times ,
and can almost spot a urei ldr in 5 seconds.
i think it would be interesting to offer the T4's in 'flavors'
urei ' what works , works'
medium ' the best of both worlds
brand x , ' when you gotta clone , a clone'
and of course my fav , ' smooth' curve. ' when it's gotta be gentle'
these just some preliminary results .
the test equipment i used is not as precise as the gear that should
arrive very soon.
the top of the graph is 1 meg ohm and the bottom start is aprox
1 k.
best .
g.
i think i should be able to work this out faily quickly.
these are the light to dark resistance curves of a batch of 30 or
so LDR's. (light dependant resistors)
interesting enough the dark to light curves seem inconsequential
for they are in the millisecond range ,
super fast ,
makes me believe that the el panel has more to do with that
then one thinks.
infact i would say the panel is a major player in this equation ,
to the point of being able to be a little loose on cell matching.\
the Brand X ldrs both land up on the red line
the UREI ldrs are more quick
alot of the LDRs are ending up more like the urei then i expected
almost dead on .
the brand x are harder to come by.
as with the 'smooth' cells
the intresing thing is now im learning attack times ,
and can almost spot a urei ldr in 5 seconds.
i think it would be interesting to offer the T4's in 'flavors'
urei ' what works , works'
medium ' the best of both worlds
brand x , ' when you gotta clone , a clone'
and of course my fav , ' smooth' curve. ' when it's gotta be gentle'
these just some preliminary results .
the test equipment i used is not as precise as the gear that should
arrive very soon.
the top of the graph is 1 meg ohm and the bottom start is aprox
1 k.
best .
g.
[quote author="dripelectronics"]made some progress using the nsl 5910's
i think i should be able to work this out faily quickly.
these are the light to dark resistance curves of a batch of 30 or
so LDR's. (light dependant resistors)
interesting enough the dark to light curves seem inconsequential
for they are in the millisecond range ,
super fast ,
makes me believe that the el panel has more to do with that
then one thinks.
infact i would say the panel is a major player in this equation ,
to the point of being able to be a little loose on cell matching.\
the Brand X ldrs both land up on the red line
the UREI ldrs are more quick
alot of the LDRs are ending up more like the urei then i expected
almost dead on .
the brand x are harder to come by.
as with the 'smooth' cells
the intresing thing is now im learning attack times ,
and can almost spot a urei ldr in 5 seconds.
i think it would be interesting to offer the T4's in 'flavors'
urei ' what works , works'
medium ' the best of both worlds
brand x , ' when you gotta clone , a clone'
and of course my fav , ' smooth' curve. ' when it's gotta be gentle'
these just some preliminary results .
the test equipment i used is not as precise as the gear that should
arrive very soon.
the top of the graph is 1 meg ohm and the bottom start is aprox
1 k.
best .
g.[/quote]
Based on the 150 or so photocells and 30 or so T4 units I've tested, I think "recovery time" is the most significant factor governing the sound of the T4 units. The next most significant factor is the spread between light and dark resistances, which is affected by the light output of the EL panel. The vintage T4A, B and C units that I've tested all fall within a fairly narrow window regarding both parameters. This weekend I'll dig out some of the old data files and post graphs on kenetek.com for comparison.
Also, I notice that your graphs only measure up to 1 Megohm. At 1 Meg, the cells are still conducting enough to provide some gain reduction. The tests that I conducted measure up to around 6.75 Megohms, which is the upper limit of the tester I built. The results are displayed on a log scale, which translates better to gain reduction, since deciBels is a logarithmic scale as well.
Also keep in mind that the photocells have significant memory after being exposed to light for even a brief time. I had to wait about 20 minutes between tests in order for the cells to recover closely enough to a "dark" state before I could get test results that could be used to compare different units. I have some NSL5910s on order from Allied that should be in before the weekend. I'll try to sneak away from family life long enough to compare those cells to the vintage T4A and T4B measurements I made almost 10 years ago and get the test results published for all to see. I'll also try to get the photos of the different T4 units (including the T4A with the paperclip as an integral part of the assembly) up on the web.
i think i should be able to work this out faily quickly.
these are the light to dark resistance curves of a batch of 30 or
so LDR's. (light dependant resistors)
interesting enough the dark to light curves seem inconsequential
for they are in the millisecond range ,
super fast ,
makes me believe that the el panel has more to do with that
then one thinks.
infact i would say the panel is a major player in this equation ,
to the point of being able to be a little loose on cell matching.\
the Brand X ldrs both land up on the red line
the UREI ldrs are more quick
alot of the LDRs are ending up more like the urei then i expected
almost dead on .
the brand x are harder to come by.
as with the 'smooth' cells
the intresing thing is now im learning attack times ,
and can almost spot a urei ldr in 5 seconds.
i think it would be interesting to offer the T4's in 'flavors'
urei ' what works , works'
medium ' the best of both worlds
brand x , ' when you gotta clone , a clone'
and of course my fav , ' smooth' curve. ' when it's gotta be gentle'
these just some preliminary results .
the test equipment i used is not as precise as the gear that should
arrive very soon.
the top of the graph is 1 meg ohm and the bottom start is aprox
1 k.
best .
g.[/quote]
Based on the 150 or so photocells and 30 or so T4 units I've tested, I think "recovery time" is the most significant factor governing the sound of the T4 units. The next most significant factor is the spread between light and dark resistances, which is affected by the light output of the EL panel. The vintage T4A, B and C units that I've tested all fall within a fairly narrow window regarding both parameters. This weekend I'll dig out some of the old data files and post graphs on kenetek.com for comparison.
Also, I notice that your graphs only measure up to 1 Megohm. At 1 Meg, the cells are still conducting enough to provide some gain reduction. The tests that I conducted measure up to around 6.75 Megohms, which is the upper limit of the tester I built. The results are displayed on a log scale, which translates better to gain reduction, since deciBels is a logarithmic scale as well.
Also keep in mind that the photocells have significant memory after being exposed to light for even a brief time. I had to wait about 20 minutes between tests in order for the cells to recover closely enough to a "dark" state before I could get test results that could be used to compare different units. I have some NSL5910s on order from Allied that should be in before the weekend. I'll try to sneak away from family life long enough to compare those cells to the vintage T4A and T4B measurements I made almost 10 years ago and get the test results published for all to see. I'll also try to get the photos of the different T4 units (including the T4A with the paperclip as an integral part of the assembly) up on the web.
Yes, the recovery curve is where you get the sound during compression.
I used to worry about the dark soak period thing, but then realized that once you hit a cell with the first note during the session, you are dealing with an "exposed" cell, unless you set the thing to just catch the peaks that come along once a day, so don't stress the dark storage thing too much, plot some recovery curves.
I use stereo VU meters, bias the cells at 0 db, then kill the light and watch the decay. It's more of a visual test, then, once you get something you like, you plug it in to make sure it passesthe audio test.
I used to worry about the dark soak period thing, but then realized that once you hit a cell with the first note during the session, you are dealing with an "exposed" cell, unless you set the thing to just catch the peaks that come along once a day, so don't stress the dark storage thing too much, plot some recovery curves.
I use stereo VU meters, bias the cells at 0 db, then kill the light and watch the decay. It's more of a visual test, then, once you get something you like, you plug it in to make sure it passesthe audio test.
[quote author="CJ"]Yes, the recovery curve is where you get the sound during compression.
I used to worry about the dark soak period thing, but then realized that once you hit a cell with the first note during the session, you are dealing with an "exposed" cell, unless you set the thing to just catch the peaks that come along once a day, so don't stress the dark storage thing too much, plot some recovery curves.
I use stereo VU meters, bias the cells at 0 db, then kill the light and watch the decay. It's more of a visual test, then, once you get something you like, you plug it in to make sure it passesthe audio test.[/quote]
That's very true, in the studio. But when comparing various photo cells (on the tester at least) I think it's important to make sure they've been dark long enough to return to their highest resistance possible. There are two reasons for this. First, recovery time is different (usually slower) after the cells have been exposed to light. Therefore, when matching cells it's important that they both be tested from a completely dark state. Second, I've had a few cells (maybe 2% or so) that I've had to reject because their dark resistance wasn't high enough, and they were attenuating slightly all the time. That's probably an anal approach but hey, I'm anal about my compressors and proud of it :grin: The T4s I built had cells that were matched easily within 1 dB and the recovery curves almost overlapped. You'll see what I mean when I get the graphs up on the web. And you can really feel a difference when using a compressor with an accurately calibrated meter. I don't know, maybe it's my imagination but somehow it makes the unit feel like it's better quality. Maybe it's just that I have more confidence in using it. YMMV.
I used to worry about the dark soak period thing, but then realized that once you hit a cell with the first note during the session, you are dealing with an "exposed" cell, unless you set the thing to just catch the peaks that come along once a day, so don't stress the dark storage thing too much, plot some recovery curves.
I use stereo VU meters, bias the cells at 0 db, then kill the light and watch the decay. It's more of a visual test, then, once you get something you like, you plug it in to make sure it passesthe audio test.[/quote]
That's very true, in the studio. But when comparing various photo cells (on the tester at least) I think it's important to make sure they've been dark long enough to return to their highest resistance possible. There are two reasons for this. First, recovery time is different (usually slower) after the cells have been exposed to light. Therefore, when matching cells it's important that they both be tested from a completely dark state. Second, I've had a few cells (maybe 2% or so) that I've had to reject because their dark resistance wasn't high enough, and they were attenuating slightly all the time. That's probably an anal approach but hey, I'm anal about my compressors and proud of it :grin: The T4s I built had cells that were matched easily within 1 dB and the recovery curves almost overlapped. You'll see what I mean when I get the graphs up on the web. And you can really feel a difference when using a compressor with an accurately calibrated meter. I don't know, maybe it's my imagination but somehow it makes the unit feel like it's better quality. Maybe it's just that I have more confidence in using it. YMMV.
Kev
Well-known member
[quote author="CJ"] ... I used to worry about the dark soak period thing, but then realized that once you hit a cell with the first note during the session, ...[/quote]
I use a tone brust or a click just before the start of a song on those channels that have this type of device
some gates would get a side chain of tone to keep them open for the song start and then the side chain is removed or bypassed
lots of old school tricks can still be used even if you have a DAW
that all gets cut off later when you trim the mixdown
I use a tone brust or a click just before the start of a song on those channels that have this type of device
some gates would get a side chain of tone to keep them open for the song start and then the side chain is removed or bypassed
lots of old school tricks can still be used even if you have a DAW
that all gets cut off later when you trim the mixdown
I've posted some pics and info on the earliest T4A and T4B units I've ever seen.
Here's the link for more pics and info: http://www.kenetek.com/OtherStuff/AboutT4units/tabid/201/Default.aspx
Here's the link for more pics and info: http://www.kenetek.com/OtherStuff/AboutT4units/tabid/201/Default.aspx
Hey, thats great!
Never seen a T4a before. That old at least.
I tried the Clairex cells right when they anounced they were not going to make them anymore.
They were a lot more dispersed, as far as on/off R and response.
So you needed to screen a lot more cells, as opposed to silonex, whose cells seem to be more consistant from batch to batch.
Never seen a T4a before. That old at least.
I tried the Clairex cells right when they anounced they were not going to make them anymore.
They were a lot more dispersed, as far as on/off R and response.
So you needed to screen a lot more cells, as opposed to silonex, whose cells seem to be more consistant from batch to batch.
[quote author="CJ"]Hey, thats great!
Never seen a T4a before. That old at least.
I tried the Clairex cells right when they anounced they were not going to make them anymore.
They were a lot more dispersed, as far as on/off R and response.
So you needed to screen a lot more cells, as opposed to silonex, whose cells seem to be more consistant from batch to batch.[/quote]
Yeah, I had a similar experience with Clairex cells. I bought 100 of them and got six good ones out of the batch. Of those six, two of them mysteriously died about 2 days after I put them in T4 units. I'm looking forward to testing the NSL5910s to see how they compare. I'll post the results ASAP.
Never seen a T4a before. That old at least.
I tried the Clairex cells right when they anounced they were not going to make them anymore.
They were a lot more dispersed, as far as on/off R and response.
So you needed to screen a lot more cells, as opposed to silonex, whose cells seem to be more consistant from batch to batch.[/quote]
Yeah, I had a similar experience with Clairex cells. I bought 100 of them and got six good ones out of the batch. Of those six, two of them mysteriously died about 2 days after I put them in T4 units. I'm looking forward to testing the NSL5910s to see how they compare. I'll post the results ASAP.
Zetajazz44
Member
Count me in for at least 1 when they're ready Greg...
Thanks in advance,
Zetajazz44
Thanks in advance,
Zetajazz44
dripelectronics
Well-known member
dripelectronics
Well-known member
gonna take some time and test a few other cells types.
when i was staring at my tes rig using an el panel from a night light ,
from CVS (all night phamacy)
i noticed this el panel had so much in common with it's 30$ counter part.
currently im testing this panel whith some light tests to gauge it's responce.
it wouldnt be a problem to make a pcb that can take 'night light panels'
should have one wired up soon , ill do a shoot out with it soon.
still waiting for the panels,
in the mean time i will start making advancements towards a
scientific vocabulary so i can have cells manufactured,
either with silonex(NSL) or one the many other cell mfg.
all the nsl 5910's range in the area of early urei and late urei models,
the relase time is quicker,
i would caution peolple who are buying t4bs for a stereo unit , and are using the NOS .
ebay t4b's ,
i opened and analized all of the cells out of four ,
the cells have a quicker recovery time then the Brand x cells ,
brand x cells are all at a 45 degree slope,
and all the urei cells have different recovery times ,
so for example , your right chanel will be slower then the left chanel
etc...
seems as though the urei testing concept was less scientific and more
speculation.
in a batch of one hundred nsl cells , i was able to match all of them give or take a handfull.
ive ordered test lots of many other cells that should arrive in the coming days ,
no need to rush this , i speculate it will be solved in a few weeks ,
the cases are ready , and wired ,
i had 'original cases manufactured , for those who want metal,
very expensive process , having dies custom made .
and for those who want the plastic cases , those will be cheaper ,
g.
when i was staring at my tes rig using an el panel from a night light ,
from CVS (all night phamacy)
i noticed this el panel had so much in common with it's 30$ counter part.
currently im testing this panel whith some light tests to gauge it's responce.
it wouldnt be a problem to make a pcb that can take 'night light panels'
should have one wired up soon , ill do a shoot out with it soon.
still waiting for the panels,
in the mean time i will start making advancements towards a
scientific vocabulary so i can have cells manufactured,
either with silonex(NSL) or one the many other cell mfg.
all the nsl 5910's range in the area of early urei and late urei models,
the relase time is quicker,
i would caution peolple who are buying t4bs for a stereo unit , and are using the NOS .
ebay t4b's ,
i opened and analized all of the cells out of four ,
the cells have a quicker recovery time then the Brand x cells ,
brand x cells are all at a 45 degree slope,
and all the urei cells have different recovery times ,
so for example , your right chanel will be slower then the left chanel
etc...
seems as though the urei testing concept was less scientific and more
speculation.
in a batch of one hundred nsl cells , i was able to match all of them give or take a handfull.
ive ordered test lots of many other cells that should arrive in the coming days ,
no need to rush this , i speculate it will be solved in a few weeks ,
the cases are ready , and wired ,
i had 'original cases manufactured , for those who want metal,
very expensive process , having dies custom made .
and for those who want the plastic cases , those will be cheaper ,
g.
jensenmann
Well-known member
[quote author="dripelectronics"]seems as though the urei testing concept was less scientific and more
speculation.
[/quote]
nice one
:grin:
You´re doing an awesome job, Gregory. Thanks a lot. This thread is really interesting.
:thumb:
speculation.
[/quote]
nice one
:grin:
You´re doing an awesome job, Gregory. Thanks a lot. This thread is really interesting.
:thumb:
Yeah, really excellent work! Count me in for four units (with metal cases) minus photocells and EL panels.
Thanks!!!
Thanks!!!
CroatianSensation
Well-known member
Join the T4 Army!!
:thumb:
:thumb:
I do not know the effects of aging on the LDR cells, this is something that takes a long time to measure.
Keep a ref cell stored, and check it a year later for response and dcr.
I think they slow down with age.
I know that the EL panel hasd a half life, so you will be turning up the GR as the units age.
Keep a ref cell stored, and check it a year later for response and dcr.
I think they slow down with age.
I know that the EL panel hasd a half life, so you will be turning up the GR as the units age.
[quote author="CJ"]I do not know the effects of aging on the LDR cells, this is something that takes a long time to measure.
Keep a ref cell stored, and check it a year later for response and dcr.
I think they slow down with age.
I know that the EL panel hasd a half life, so you will be turning up the GR as the units age.[/quote]
From my experience the cells slow down both with age and prolonged exposure to heat.
Keep a ref cell stored, and check it a year later for response and dcr.
I think they slow down with age.
I know that the EL panel hasd a half life, so you will be turning up the GR as the units age.[/quote]
From my experience the cells slow down both with age and prolonged exposure to heat.
Similar threads
